We present results two years after the end of a group-based parenting intervention tested in a cluster randomized control trial in rural Kenya. The original program consisted of 16 fortnightly village-based sessions over 8 months and had large positive impacts on children's cognition and parenting behaviors immediately after its end. Over the next two years, a random half of intervention villages received a light-touch "booster" intervention to offer continued yet less intensive program support. With and without the booster extension, early program impacts were sustained two years later, albeit smaller in magnitude. Boosters had a small positive added value on parenting behaviors and children's socioemotional development, despite the interruption of COVID-19 to their delivery. Sustained impacts on children's development were strongly mediated by improvements in parenting behaviors, disadvantaged families accrued the largest benefits, and two years later our program remains one of the most cost-effective and potentially scalable programs globally to date. These results point to encouraging paths forward for maximizing the reach and longer-term effectiveness of early childhood development programs to improve child development in low-resource remote settings.